# RIDM: Reinforced Inverse Dynamics Modeling for Learning from a Single Observed Demonstration

## TL;DR

RIDM introduces a novel method that combines imitation from observation with reinforcement learning, requiring only a single demonstration and no action data, demonstrating strong performance in simulation and real robot tasks.

## Contribution

RIDM is a new paradigm that learns from a single demonstration without needing demonstrator actions, bridging imitation from observation and reinforcement learning.

## Key findings

- RIDM outperforms baseline methods in simulation tasks.
- RIDM successfully applied to real UR5 robot arm tasks.
- RIDM operates effectively with raw state features and a single demonstration.

## Abstract

Augmenting reinforcement learning with imitation learning is often hailed as a method by which to improve upon learning from scratch. However, most existing methods for integrating these two techniques are subject to several strong assumptions---chief among them that information about demonstrator actions is available. In this paper, we investigate the extent to which this assumption is necessary by introducing and evaluating reinforced inverse dynamics modeling (RIDM), a novel paradigm for combining imitation from observation (IfO) and reinforcement learning with no dependence on demonstrator action information. Moreover, RIDM requires only a single demonstration trajectory and is able to operate directly on raw (unaugmented) state features. We find experimentally that RIDM performs favorably compared to a baseline approach for several tasks in simulation as well as for tasks on a real UR5 robot arm. Experiment videos can be found at https://sites.google.com/view/ridm-reinforced-inverse-dynami.